Improvement in Computational Complexity of the MIMO ML Decoder in High Mobility Wireless Communicati

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 04 Issue: 09 | Sep -2017

p-ISSN: 2395-0072

www.irjet.net

Improvement in computational complexity of the MIMO ML Decoder in high mobility wireless communication systems using non-stationary IMT-A K.Krishna Kiriti Sai1 , Sk.Shakeera2 1(M.

Tech, ECE Department, VR Siddhartha Engineering College, Kanuru, Vijayawada professor ECE Department, VR Siddhartha Engineering College, kanuru, vijayawada ---------------------------------------------------------------------***--------------------------------------------------------------------2(Assistant

Abstract - The MIMO has attracted a lot of attention recently as a promising technique that can enhance the system performance and link reliability. The high mobility wireless communication faces the problem of providing the reliable communication over rapidly varying radio channels.

Measurements have demonstrated that the standard channel models, like IMT-A,WINNER-II models offer stationary intervals considerably larger than high mobility wireless systems. The stationary interval defined, maximum time duration over which channel satisfy the WSS condition. In this project a non-stationary model with time varying parameters like number of clusters, angle of departures, angle of arrivals power and delays of angle of arrivals along with the statistical parameters like spatial cross correlation function and temporal auto correlation function will be discussed. At the receiver, the signal considered, is from the multiple transmits antennas, as a vector. The traditional search method determines component in the transmitted vector, one by one. It is known that order of decisions on signal components has large impact on the computational complexity of the search part. QR decomposition with sort, determines an efficient order of decisions on signal components, decreases the computational complexity of the search part with little increase of computational complexity in pre-process part. The search part can be regarded as shortest path problem in a weighted graph and use of the Dijkstra’s algorithm significantly reduces the computational complexity of the search part.

Keywords - IMT-Advanced, MIMO, Non-Stationary, time varying parameters, statistical parameters, Dijkstra’s Algorithm.

INTRODUCTION MIMO technology has attracted the lot of attention recently as a promising technique that can enhance the system performance. The channel has a noteworthy part in deciding the unwavering quality of the method. In high mobile wireless communication the data transmission is of high data rates, the large amount of data at the receiver, it makes very difficult for the receiver to isolate the each user data which is having the high accuracy.

Due to this, the data corruption and the reliability of the channel get effected. In order to overcome this problem the amount of data has to be minimized by the help of some technique. QR Decomposition is technique. QR decomposition technique is used in a way to reduce the complexity at the receiver by following some steps. QR decomposition with sorting further reduces the complexity at the receiver. An algorithm called Dijkstra’s algorithm find the shortest path from one source to the destination. This results in reduced complexity at the receiver.

QR DECOMPOSITION AND SORTING In this technique the signal travels through the channel. The faded signal matrix is received at the receiver. Since the received signal is in the vector form it undergoes into conversion as matrix. A standard low pass equivalent is generated as

H as fading channel matrix, is transmitted signal is additive white Gaussian noise, S is signal constellation. Let H=QR be QR decomposition, be an ML estimate of .Let ||

||=||

=||Q* -Q*QR|| =||Q* -R || is the minimum value among all transmitted signals in .

We will formally review the original method in search part[].It tries to find a vector within the sphere of radius

and the centre

and considers the in equality condition ||Q* -R || C. The t-th column in Q* -R is qt+ So we choose

© 2017, IRJET

|

Impact Factor value: 5.181

|

||

∈ S with |qt -

(1) and

is necessary condition for (1). |2 ≤ C as a candidate of t-

ISO 9001:2008 Certified Journal

|

Page 804

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 04 Issue: 09 | Sep -2017

p-ISSN: 2395-0072

th component in

www.irjet.net

Considering (t - 1)-th and t-th columns in

Q∈ - R , we choose

∈ S with

output finds the shortest path from source to destination.

The only thing that increased here is the space for processing.

RESULTS + as a candidate of (t - 1)-th component in . Remaining is resolved in a comparable manner. If there is no candidate ∈ S, then we rechoose .When we determine ( ), we set the new radius to | | ||and repeat the search part. The last found vector is the ML estimate of the transmitted signal. On the off chance that there is no conceivable transmitted flag in the circle with starting span c then we announce eradication or increment the range c

The final result is to find the shortest path from source to destination point. The shortest path is found for the calculated signal output. Scatter plot is the first result from where the scatter outputs are calculated after passing the Rayleigh fading channel. Performing the QR decomposition and the sorting, the output is applied for the algorithm process. The output from algorithm is represented in graph.

The shortest path values for the calculated scattered output are represented..

Scatter Plot

Need for Sorting The sorting is process of arranging the items systematically.

Ordering and categorizing. With this procedure we get a qr disintegration of the segment permuted network of of H. In the event that we apply the customary inquiry part in segment 2 with then we get all the more proficiently the permuted variant of the ml appraise . The ML estimate can be obtained by the permutation.

QR DECOMPOSTION WITH SORTING We will describe the QR decomposition with sort.

Watch that in the conventional qr decay is equivalent to the standard of the principal section vector of H. In order to minimize , we replace the first column of H with the column with minimum norm. Let H be the column replaced version of H .Register a unitary lattice with the end goal that the main section of is the of Q1H is . Let be (r - 1) × (t - 1) submatrix of Q 1H with the first column and the first row of Q 1H removed. Replace the first column of with the column with minimum norm in . Let be the column replaced version of . Compute a unitary matrix such that the first column of is .The Calculation process is recursively rehashed until i =t

Fig:-Scatter Plot of the Input Signal

DIJKSTRA’S ALGORITHM Dijkstra’s algorithm is an algorithm used for finding the shortest paths between the nodes in a graph. In high mobility wireless communication system, the information accumulated at the receiver is very large. In order to satisfy the needs of the user the huge information is partitioned by using the QR Decomposition with sort. The information is pre-processed in the decomposition process and the sorting; the processing part is further reduced. The computational complexity is reduced. The use of algorithm for the sorted

© 2017, IRJET

|

Impact Factor value: 5.181

|

ISO 9001:2008 Certified Journal

|

Page 805

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 04 Issue: 09 | Sep -2017

p-ISSN: 2395-0072

www.irjet.net

Shortest Path Values

© 2017, IRJET

|

Impact Factor value: 5.181

|

ISO 9001:2008 Certified Journal

|

Page 806

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 04 Issue: 09 | Sep -2017

p-ISSN: 2395-0072

www.irjet.net

CONCLUSION The shortest path distance form source node to the destination node can be found using this method in high mobility wireless communication system. In order to achieve this some exemptions are considered like more space for processing. In future the shortest path can be achieved by minimizing the space complexity.

References [1] Ammar Ghazal, Student Member, IEEE, Yi Yuan, ChengXiang Wang, Senior Member, IEEE, Yan Zhang, Member, IEEE, Qi Yao, Hongrui Zhou, and Weiming Duan, “A NonStationary IMT-Advanced MIMO Channel Model for HighMobility Wireless Communication Systems” [2] Takayuki FUKATANI†, Ryutaroh MATSUMOTO‡, and Tomohiko UYEMATSU,”Two Methods for Decreasing the Computational Complexity of the MIMO ML Decoder”, Dept. of Communications and Integrated Systems, Tokyo Institute of Technology, 152-8552 Japan.

[3] V. Erceg, et al., IEEE 802.11 document 03/940r4, “TGn channel models, ”May 2004 [4] A. V. Aho, J. E. Hopcroft, and J. D. Ullman. Data Structures and Algorithms. Addison-Wesley, 1983.

© 2017, IRJET

|

Impact Factor value: 5.181

|

ISO 9001:2008 Certified Journal

|

Page 807